## Reading layer `landkreise-in-germany' from data source `C:\Users\nauss_lokal\Documents\plygrnd\CovidAirPolution\data\DE\landkreise-in-germany\landkreise-in-germany.shp' using driver `ESRI Shapefile'
## Simple feature collection with 403 features and 14 fields
## geometry type:  MULTIPOLYGON
## dimension:      XY
## bbox:           xmin: 5.866251 ymin: 47.27012 xmax: 15.04182 ymax: 55.05653
## CRS:            4326
##  [1] "LK Böblingen"                "LK Breisgau-Hochschwarzwald"
##  [3] "LK Erzgebirgskreis"          "LK Esslingen"               
##  [5] "LK Ludwigsburg"              "LK Mansfeld-Südharz"        
##  [7] "LK Nordfriesland"            "LK Oberhavel"               
##  [9] "LK Schwarzwald-Baar-Kreis"   "LK Uelzen"                  
## [11] "SK Freiburg i.Breisgau"      "SK Heidelberg"              
## [13] "SK Heilbronn"                "SK Pforzheim"               
## 
##  Precomputing distance matrix...
## 
## Iteration 1: Changes / Distsum =  198 / 21618
## Iteration 2: Changes / Distsum =   25 / 14947
## Iteration 3: Changes / Distsum =    9 / 14620
## Iteration 4: Changes / Distsum =    8 / 13278
## Iteration 5: Changes / Distsum =    9 / 11590
## Iteration 6: Changes / Distsum =    2 / 11561
## Iteration 7: Changes / Distsum =    0 / 11561
## 
##  Elapsed time is 0.16 seconds.
## 
## 
##  Precomputing distance matrix...
## 
## Iteration 1: Changes / Distsum = 198 / 37653.5
## Iteration 2: Changes / Distsum = 62 / 31797.16
## Iteration 3: Changes / Distsum = 11 / 31428.25
## Iteration 4: Changes / Distsum =  9 / 31285.6
## Iteration 5: Changes / Distsum =  0 / 31285.6
## 
##  Elapsed time is 0.15 seconds.

Results for PM

## [1] "Start date for analysis:  2020-02-15"
## [1] "End date for analysis:    2020-04-01"
## [1] "PM values:                PM10"

Countrywide temporal development of PM values and COVID-19 cases in Germany

Dynmiac time warp clustering on countrywide average

Explanatory potential of PM2.5 for COVID-19 cases

Hier mal ein Versuch, die “Italien-Grafik” aus dem etwas nachzubauen. Die Signifikanz ist bei PM10 für längere Zeitintervalle gegeben, für PM2.5 praktisch nicht (Setti, L. et al. The Potential role of Particulate Matter in the Spreading of COVID-19 in Northern Italy: First Evidence-based Research Hypotheses. medRxiv, https://www.medrxiv.org/content/10.1101/2020.04.11.20061713v1 (2020).)

## 
## Call:
## glm(formula = test_merge$ratio ~ test_merge$pm_gt_10, family = quasipoisson)
## 
## Deviance Residuals: 
##        Min          1Q      Median          3Q         Max  
## -0.0014353  -0.0007323  -0.0004128   0.0002041   0.0060236  
## 
## Coefficients:
##                      Estimate Std. Error t value Pr(>|t|)    
## (Intercept)         -15.32370    0.25852 -59.274  < 2e-16 ***
## test_merge$pm_gt_10   0.07161    0.01425   5.025 1.13e-06 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## (Dispersion parameter for quasipoisson family taken to be 1.374506e-06)
## 
##     Null deviance: 0.00019705  on 197  degrees of freedom
## Residual deviance: 0.00016314  on 196  degrees of freedom
## AIC: NA
## 
## Number of Fisher Scoring iterations: 16